The assessment of low-level contamination from gamma-emitting radionuclides

Directorate-General For Research And Innovation European Commission Directorate-General For The Information Society And Media

Description

Nuclear energy and safety
Energy research

Informations

• Publié par: DIRECTORATE-GENERAL-FOR-RESEARCH-AND-INNOVATION-EUROPEAN-COMMISSION-DIRECTORATE
• Nombre de lectures: 5
• Langue: English
• Poids de l'ouvrage: 6 Mo

Extrait

Commission of the European Communities
The assessment of low-level contamination
gamtnä-€mitting from radionuclides
P. \/V. Gray
lmperial College Reactor Centre
Silwood Park, Ascot
Berkshire 7PY, SLs United Kingdom
Research contract No FllD-0019
Final report
performed Work in the framework the of European Community's
research programme'Decommissioning nuclear of installations'
Project No 6'Estimation the quantities of of radioactive waste arising from decommissioning
of nuclear installations in the Community'
Directorate-General
Science, Research and Development
1 989 EUR 121 83 ENPublished by the
OF THE EUROPEAN COMMUNITIESCOMMISSION
Directorate-General
InnovationTelecommunications, Information lndustries and
L-2920 Luxembourg
NOTICELEGAL
person Communities nor any actingNeither the Commission of the European
made ofis responsible for the use which might be on behalf of the
the following information
this publicationCataloguing data can be found at the end of
Communities, 1989Luxembourg: Office for Official Publications of the European
number: CD-NA-12183-EN-CCatalogue ISBN 92-825-9994-9
1989ECSC-EEC-EAEC, Brussels' Luxembourg, @
in BelgiumPrinted ABSTR,ACT
This report describes a collection of techniques ttrat can be used to make inferences about
peak specttal area, and hence about source activity, for radionuclide specfa in general, and
for the spectra of gamma-emiuing particular. radionuclides in These techniques for spectral
analysis:
- provide accurate peak arera uncertainties and confidence intervals for poorly defined
- spearat peaks spectral peaks for which the least squares ard maximum likelihood
methods prove unrcliable;
- -allow the precision with which peak spectral area can be esimated to be increased
an increase of up to 85% free parameter over least squarcs or maximum likelihood
estimation;
- permit peak tlrc cennoid separation peaks at which overlapping spectral can be resolved
- to be decreased a decrcase of up ftee parameter to 9l% over least squar€s or
maximum likelihood estimation;
- extend the circumstances in which peak area limits hypothesis detection and tests can
be constnrcted to include those cases wherc a spectral pealc with overlapa one or more
qpectral peaks;other
- offer lower a detection limit for radionuclide activity than üat cunently available wittr
- gross-counting a decrease of 28% in the detection limit, and 48% a decrease of n
the counting time;
can be used to calculate the extent to which üesting simultaneously for tlre presence of
a number of radionuclides within a sample lowers the detection limit associated with
each radionuclide:
- provide probabilistic an empirical model for peak the urrcertainties in the shape and
location pal'ameters arising fiom calibration wrcertainties ard spectral in drift a
spectfometry system;
- can be used increase to the rate of convergence and the robustness of a peak-fining
algorittim.
-il|TABIE OF CONTENTS
Page
ABSTRACT
TABLE OF CONTENTS
r. OBJECTIVES
trsM 5
1. Methodology for peak-fitting 7
2. Results 10
m. 21
1. Intensities and energies 23
2. Inferences peak about spctral arca 29
3. Pre-acquisition Bayesian prior probability density functions 36
4. Estimation and confidence intervals 56
5. Hlpothesis üesting and detection limits 73
fv.w 99
V. APPENDD( 103
l. The representation of measurement uncenainty 105
2. Hlpottrcsis testing 114
3. he-acquisition prior Bayesian probability density functions 119
4. General linear model 128
Notatim
Numbered rcferences to literature items arc contained within square brackets. Numbered
'Methodology'refercnces to parts of (in this report round brackets) refer to the division
unless otherwise stated. References (denoted within a division are divided iffio sections by
'S') 'ss'). the letter and to subsections (denoted (S5.2-ss3.2.3) by the letters Thus denotes
subsection 3.2.3 of section 5.2.r. OBJECTITYES